Database systems are commonly employed by organizations to store data. Accordingly, the database systems may have to be reliable and scalable. For reliability, in a distributed environments, the stored data may be replicated from a primary data center to a secondary data center. The secondary data center can replace the primary data center without data loss upon unavailability of the primary data center. When the data is updated at the primary data center, the secondary data center is updated to reflect the changes to the data. However, replicating data such as the data specific to the primary data center, data which can be rebuild and the like may not be useful and such process may only increase network traffic and overload the secondary data center.